The Divergence of Major Scale Insect Lineages (Hemiptera)
Total Page:16
File Type:pdf, Size:1020Kb
www.nature.com/scientificreports OPEN Putting scales into evolutionary time: the divergence of major scale insect lineages (Hemiptera) Received: 14 October 2015 Accepted: 08 March 2016 predates the radiation of modern Published: 22 March 2016 angiosperm hosts Isabelle M. Vea1,2 & David A. Grimaldi2 The radiation of flowering plants in the mid-Cretaceous transformed landscapes and is widely believed to have fuelled the radiations of major groups of phytophagous insects. An excellent group to test this assertion is the scale insects (Coccomorpha: Hemiptera), with some 8,000 described Recent species and probably the most diverse fossil record of any phytophagous insect group preserved in amber. We used here a total-evidence approach (by tip-dating) employing 174 morphological characters of 73 Recent and 43 fossil taxa (48 families) and DNA sequences of three gene regions, to obtain divergence time estimates and compare the chronology of the most diverse lineage of scale insects, the neococcoid families, with the timing of the main angiosperm radiation. An estimated origin of the Coccomorpha occurred at the beginning of the Triassic, about 245 Ma [228–273], and of the neococcoids 60 million years later [210–165 Ma]. A total-evidence approach allows the integration of extinct scale insects into a phylogenetic framework, resulting in slightly younger median estimates than analyses using Recent taxa, calibrated with fossil ages only. From these estimates, we hypothesise that most major lineages of coccoids shifted from gymnosperms onto angiosperms when the latter became diverse and abundant in the mid- to Late Cretaceous. Living insect species that feed on vascular plants comprise some 40% of the described insect diversity1, and so it appears that plants have had a profound effect on the diversification of insects. In comparisons between multiple sister-pairs of insect groups, for example, where one group is herbivorous and the other not, the former was found to be almost always far more diverse2. Moreover, within major groups of herbivorous insects the great propor- tions of species feed on angiosperms, the sister lineages having just a few species that feed primarily on gymno- sperms or plant detritus. Good examples include the hyperdiverse weevils (Coleoptera: Curculionoidea)3 and the Lepidoptera1,4, the latter the largest lineage of plant-feeding animals. Since both of these insect groups are known to pre-date the first fossil angiosperms, and well preceded the angiosperm radiations in the mid-Cretaceous, it has commonly been inferred that the “colonization” or shift to angiosperms promoted insect diversification. Perhaps no other group has been a more popular subject for this topic than butterflies (Lepidoptera: Papilionoidea), in which diversification of the whole group and component lineages have been tied to the diversification of various host plant lineages5–8. Without question, the pervasiveness of insect herbivory has been a major selection pressure on plants, which evolved pharmacopias of toxic secondary metabolites to defend against the insects5,9, and insects in turn evolved resistance to the toxins7. However, the role that herbivory per se has played in insect diversification, speciation, or cladogenesis, is unclear. First, the 280,000 known angiosperms10 constitute the largest majority of the living vascular plant diversity, so a preponderance of insect herbivores on angiosperms would be expected based on chance alone. Second, there was no apparent increase in the number of insect families during the Cretaceous 1Richard Gilder Graduate School, American Museum of Natural History, Central Park West at 79th street, New York, NY 10024, USA. 2Division of Invertebrate Zoology, American Museum of Natural History, Central Park West at 79th street, New York, NY 10024, USA. Correspondence and requests for materials should be addressed to I.M.V. (email: [email protected]) SCIENTIFIC REPORTS | 6:23487 | DOI: 10.1038/srep23487 1 www.nature.com/scientificreports/ Figure 1. Representatives of Coccoidea showing extreme sexual dimorphism: neococcoid families. (A) Pseudococcidae adult female (center) and male (small on female, left of ant) Phenacoccus sp., credit: Sergio Jansen González, (B) Coccidae adult female (left) and male (right) Pulvinaria acericola (Walsh & Riley), credit: Matt Bertone, (C) “archeococcoid” family: Ortheziidae adult female (top) and male (bottom) Praelongorthezia praelonga (Douglas), credit: Mark Yokoyama, (D) inclusion of Kozarius perpetuus (Vea & Grimaldi) in mid- Cretaceous Burmese amber (~100 Ma), (E) inclusion of Hodgsonicoccus patefactus (Vea & Grimaldi) in Lebanese amber (~130 Ma). when angiosperms radiated11. Third, there are very few definitive examples with appreciable correspondence in insect-host plant relationships12, exceptions being groups like fig wasps (Hymenoptera: Agaonidae)13 and yucca moths (Lepidoptera: Prodoxidae), whose larvae feed on the same hosts for which the adult insects are also spe- cialized pollinators14. Fourth, a few studies done to date indicate that some lineages of herbivorous insects actu- ally diversified after their host taxa15. Lastly, and which would explain the general lack of co-speciation between insects and their host plants, a population-genetic mechanism for how host-plant use would lead to sympat- ric divergence has been empirically controversial and largely unproven16,17. We explored the diversification of a major group of phytophagous insects, the scale insects (Hemiptera: Coccomorpha), of which 97% of the living species feed on angiosperms, and the group as a whole has a superb fossil record for the past 130 million years. Coccomorpha are Hemiptera, all of which possess mouthparts modified into a rostrum, comprised of highly specialized mouthpart appendages that allow them to pierce and siphon liquids, from plant vascular fluids to insect hemolymph and vertebrate blood. This feature not only adapted most hemipterans to an exclusive diet of plant fluids (90% of them1), but as a consequence some members have developed intimate symbiotic relation- ships, such as with ants, which feed on their excreted honeydew18 and with endosymbionts that nutritionally supplement a diet of plant fluid19. Within the strictly phytophagous suborder Sternorrhyncha (also including whiteflies, plant lice and aphids), the most speciose infraorder is Coccomorpha20 (scale insects and mealybugs), representing half of the species diversity and accounting for some of the most important plant pests. There are some 8,000 described species of Coccomorpha21, with some 52 families (33 Recent and 19 extinct); the sis- ter group, Aphidomorpha, comprises three Recent families with about 4,500 species22. Despite the impact of scale insects on agriculture and an extensive body of taxonomic work, very few studies have addressed their evolutionary history, which clearly impedes evolutionary understanding of relationships to their host plants. Higher-level phylogenetic relationships are gradually becoming better resolved23,24, diverse new fossils are being uncovered25–28, allowing timelines of lineage divergence to be assessed. The recognized monophyletic lineage neo- coccoids24,29 constitutes 90% of Coccomorpha Recent species (e.g., Pseudococcidae (Fig. 1A), Coccidae (Fig. 1B), and Diaspididae) and comprises half of the families. In contrast, the remaining scale insect families, comprising the informal, basal paraphyletic grade “archeococcoids” (Fig. 1C), are significantly less diverse today, with only 10% of the species. How did the neococcoids become so diverse today? Most scale insects feed on angiosperms24, with some exceptions amongst the archeococcoids, such as the Matsucoccidae (ca. 30 Recent species, exclusively feeding on SCIENTIFIC REPORTS | 6:23487 | DOI: 10.1038/srep23487 2 www.nature.com/scientificreports/ conifers), or the Ortheziidae (ca. 200 Recent species, mostly found in leaf litter or on lichens and mosses30). Thus, the straightforward hypothesis is that neococcoid families originated and diversified as a consequence of the angi- osperm radiations in the mid-Cretaceous31,32. Scale insects have an exemplary fossil record for the past 130 Ma, since they are one of the most abundant and diverse groups preserved in amber deposits around the world25–27, from the Early Cretaceous to the Miocene (ca. 130–20 Ma). In Turonian-aged (90 Ma) amber from New Jersey, for example, Coccomorpha is the most abundant family-level group of insects33. Moreover, the microscopic fidelity of preservation in amber allows rigorous phylogenetic interpretation of the inclusions, particularly of such min- ute insects (Fig. 1D,E). Based on recent discoveries of new coccoids in Cretaceous amber and their phylogenetic relationships28, which we further explore here, the view of a Cretaceous radiation of Coccomorpha and Tertiary radiation of neococcoids needs to be revised. Because of extreme sexual dimorphism in coccoids (Fig. 1A–C), the conspicuous, feeding, and longer-lived adult females are used in taxonomy. In contrast, the ephemeral, winged males are devoid of mouthparts and emerge for reproduction exclusively. For a large majority of coccoid genera and species the male is unknown. Although not an issue for species identification and delineation, this is problematic for phylogenetic studies since the highly reduced, paedomorphic females dramatically differ among families (e.g., Fig. 1A–C). Homologies are obscured, resulting in little phylogenetic study of female morphology and a reliance on molecular